National Centre of Competence in Research PlanetS
Sidlerstrasse 5 | CH-3012 Bern | Switzerland
  Tel. +41 (0)31 631 32 39
Fax. +41 (0)31 631 44 05

Characteristics of planets: Internal structure and atmosphere

Project 5.2 concerns the characterisation of planets and exoplanets, which is the elucidation of the physics and chemistry of their atmospheres and interiors. The work packages include WP5 (synthetic spectra of atmospheres), WP6 (atmospheric chemistry) and WP7 (interior structure and interior-atmosphere exchanges). We are interested in the processes and mechanisms governing the radiation, chemistry and dynamics of (exo)planetary atmospheres and their interiors, as well as the obstacles and degeneracies associated with interpreting astronomical data and the lessons we may learn from the Solar System.

For WP5, Ph.D student Baptiste Lavie (M.S., Paris) is shared between Projects 3 and 5 and works on applying inversion techniques to interpreting spectra of exoplanetary atmospheres. Specifically, these “atmospheric retrieval” methods utilise modern techniques (Markov Chain Monte Carlo, optimal estimation, etc) and are able to compute posterior distributions of quantities that have uncertainties proportional to the data quality. In other words, our inferences on the chemical abundances and thermal structure of an atmosphere are only as good as the error bars on the data. Lavie has just mastered the optimisation technique and has successfully tested it on a single-parameter blackbody function. The next steps are to build a forward model for a one-dimensional atmosphere and link the optimisation technique to the calculation of synthetic spectra. Our novel contribution will be to include the effects of clouds/aerosols formally into the optimisation technique, an important point that has been overlooked in the literature. We anticipate that Lavie’s work in Project 5 will be highly relevant to the data measured by Project 3.

For WP6, Ph.D student Shang-Min Tsai (M.S., National Taiwan University) works on constructing generic chemical networks with applications to exoplanetary and proto-planetary atmospheres. Tsai has just succeeded in using an inverse Euler method to construct an equilibrium chemistry network involving just hydrogen and oxygen. The next steps are to add carbon, sulphur and phosphorous, as well as generalising the procedure to consider disequilibrium thermochemistry (out-of-equilibrium chemistry due to mechanical motion of the atmosphere). Eventually, we will also consider photochemistry: disequilibrium chemistry induced by illumination by ultraviolet photons. These chemical networks may also be adapted and tabulated to work in tandem with three-dimensional climate models of exoplanets. Numerical improvements will also be sought: the inverse Euler method may be replaced by the Rosenbrock method, which is the standard workhorse for solving stiff sets of equations. We anticipate that Tsai’s work will have a broad relevance towards interpreting and understanding data from Project 3.

For WP7, Dr. Frank Wagner (Ph.D, Berlin) has just been hired. He will compute the dynamics of the interior of terrestrial to super-Earth type planets. This WP will build on the extensive expertise of Prof. Paul Tackley’s group in two- and three-dimensional modeling of the long-term evolution of the crust, mantle and core of Earth and various other terrestrial planets. The thermal structure of the atmosphere and tides raised by the star will be explicitly considered. Using these models, we will compute the level of outgassing of volatiles in these planets, as well as the amount of volatiles recycled into the interior (e.g., water, carbon dioxide). These two effects are particularly important, because they influence the atmospheric structure. Finally, these models, presently developed mainly for volatile-poor planets, will be extended during the NCCR to volatile-rich planets, which will allow us to study and predict the interior structure of putative “ocean planets” and consider the internal structure of the regular Jovian satellites; the latter will have relevance for the JUICE mission.


Preliminary tests of our atmospheric retrieval algorithm, where we attempt to invert for the temperature of a blackbody function using a curve with artificial noise inserted.  The Monte Carlo method requires a “burn-in” phase, during which outputs should be disregarded, and computes the full posterior distribution of the blackbody temperature.  Eventually, we will extend this method to inverting a family of parameters from spectra of exoplanetary atmospheres.  (Credit: Baptiste Lavie)



Relative abundances of various atoms and molecules as computed using a custom-built chemical reaction network that solves for chemical equilibrium using a time-dependent method (rather than a relaxation/Newtonian method).  The abundances are normalised relative to atomic hydrogen.  At the temperatures considered, hydrogen exists predominantly in its atomic form.  We consider a system only with hydrogen and oxygen.  At the high temperatures considered, atomic oxygen, water and the hydroxyl radical are the important species that emerge; molecular oxygen and ozone have negligible abundances at these temperatures. (Credit: Shang-Min Tsai)